Survival and Growth Factors Affecting Community-Planted Urban Street Trees

This Article is brought to you for free and open access by the Biology at Digital Commons at Loyola Marymount University and Loyola Law School. Ithas been accepted for inclusion in Cities and the Environment (CATE) by an authorized administrator of Digital Commons at Loyola MarymountUniversity and Loyola Law School. For more information, please contactdigitalcommons@lmu.edu.

, energy conservation (Nowak 1995), wildlife habitat (Coder 1996), weather 44 buffering and urban heat island amelioration (McPherson 1994), storm water catchment trafficked streets, in yard trees, in trees without wounding, and in larger pit cuts.Species-specific 99 variability in DBH and crown fullness is also expected based on tree habit and form.Japanese tree lilac (Syringa reticulata (Blume) H.Hara, Oleaceae).Records for these ten species 132 were sorted by stock and year planted, and then randomized.Up to 30 sites for each stock, of 133 each species, from each planting year, were randomly selected (some groupings were limited to 134 fewer than 30 sites).The sampling design was fully crossed, and blocked by planting year.

135
Because some neighborhoods are more involved in TreeVitalize plantings than others, sites were 136 not distributed evenly across the city.A total of 1411 sites were selected through this process and 137 mapped using ArcGIS (Figure 1).Of the original 1411 trees sites, 644 (45%) were B&B trees, 138 and 767 (55%) were bare root trees.were verified or revised.Three growth/vigor measures were then taken: diameter at breast height 148 (DBH), percent canopy cover, and percent live crown.DBH was measured in quarter-inch 149 increments using calipers.Four measurements of percent canopy cover were taken using a 150 densiometer.These measurements were taken at curbside, left, right, and sidewalk-facing 151 directional points; and averaged to capture variability.In order to calculate percent live crown, 152 trunk height and total height were measured in half-foot and foot increments, respectively.A two-way ANOVA with an error term for years planted (age) was also done to test for an interaction between species and stock while accounting for the blocked sampling design (Peña-171

174
A majority of the 1411 tree sites selected were found and inspected (1159, 82.6%).Those 175 not found were either, a) never planted, or b) planted, died and removed.89% of the B&B sites 176 and 77% of bare root trees were found.While this suggests a potentially higher rate of mortality 177 in bare root trees, in reality this discrepancy is due to a higher rate of clerical mistakes in early 178 bare root planting years.Sites not found were not included in statistical analysis.Years since planting ("age") were also not a significant predictor of mortality (p=0.921).

189
Species with the highest survival rates were P. virginiana and Platanus x acerifolia, with A.
INTRODUCTION 39 Urban forests have been recognized for their role in improving the standard of living for 40 city residents across the country (Dwyer et al. 1992; Bolund & Hunhammer 1999; Sather et al. 41 2004).Street trees offer a wide array of services including improved air and water quality 42 (Beckett et al. 2000; Nowak et al. 2007), property value (McPherson et al. 1997), human health 43

103
This study was conducted on trees planted in Philadelphia through the southeast 104 Pennsylvania TreeVitalize campaign coordinated by the Pennsylvania Horticultural Society 105 (PHS).Philadelphia is located at 39˚ 57' 8" N / 75˚ 9' 51" W along the mid-Atlantic border of 106 the United States.The city covers 326.14 km 2 and is situated at an elevation of 11.89 m above 107 mean sea level.It is home to 1.5 million people and 2.1 million trees.The city has an average 108 canopy cover of 15.7% (ranging from 1.8% to 38.3% by neighborhood) (American Forests 109 2003), and more than half of its trees have diameter at breast height (DBH) sizes of less than 110 15.25 cm (6 in) (Nowak et al. 2007).Every year hundreds of trees are planted by the city 111 government through the Department of Parks and Recreation (through a division formerly called 112 the Fairmount Park Commission).Many hundreds, and in recent years thousands, more are 113 planted by PHS through the TreeVitalize campaign.114 TreeVitalize is a public-private partnership launched by the PA Department of 115 Conservation and Natural Resources in 2004.In less than five years the program reached its 116 initial goal of planting over 20,000 trees in and around Philadelphia through community 117 members.The program continues to grow in southeast PA and has now been launched in all 118 other metropolitan areas across the state.Tree-planting volunteers are led by community group 119 leaders trained through the 9-hour Tree Tender® course developed by PHS in collaboration with 140

Figure 1 :
Figure 1: Based on records from the Pennsylvania Horticultural Society, 1411 bare root

145
Site inspections were conducted between mid-June and early August to ensure that all 146 trees were fully leafed-out upon time of inspection.At each site, address and species planted 147 Philadelphia Street Tree Sample Sites: Bare Root and B&B trees

153
Measurements followed Colorado State Forest Service guidelines(Schomaker 2004).Percent 154 live crown was later calculated by dividing live crown height (total height -trunk height) by 155 total height and multiplying by 100.Street traffic intensity (residential low traffic, residential 156 high traffic, or commercial) was based on visual assessment of site land use and traffic intensity 157 (both vehicular and pedestrian).Site type (commercial corridor, curbside, yard, park), tree trunk 158 wound presence, and sidewalk pit cut dimensions were also recorded.159 Statistical Analysis 160 Statistical analysis was done using R statistical software.DBH data was normalized using 161 a log transformation, while percent canopy cover and percent live crown were normalized using 162 an arcsin() 2 transformation.Logistic regression was done to analyze mortality data (Packer and 163 Clay 2000), and multiple linear regressions in conjunction with regression trees were used to 164 analyze growth measure data (DBH, percent canopy cover, percent live crown) (Gregg et al. 165 2003).Regression trees were used as a visual aid in determining significant interaction effects 166 (De'ath and Fabricius 2000).Explanatory variables included stock, species, age, site traffic, site 167 type, wound presence, and pit cut size.I used a forward selection procedure to retain only those 168 factors that were significant in my model (Peña-Claros et al. 2008, De'ath and Fabricius 2000). 169 and B&B stocks exhibited very high survival rates: 96% of B&B trees, 181 and 95% of bare root trees after an average of 2.62years since planting.The results of logistic 182 regression analysis indicated no significant difference in survival rates between bare root and 183 B&B trees (Figure 2).

184Figure 2
Figure 2High rates of survival were found both stocks; with no significant difference in survival 190 ginnala, S. reticulata, Amelanchier spp., and C. kentukea close behind; C. canadensis had the 191 lowest survival rate (Figure 3).

198 199
Street traffic intensity was the only other significant factor to affect survival rates.Lower 200 survival rates were observed along heavily trafficked commercial corridors (p=0.0056) (

Growth 207 Growth
was evaluated using three measures: diameter at breast height (DBH), percent 208 canopy cover, and percent live crown.Due to discrepancies between recorded and actual species 209 planted on sites, only Acer rubrum (red maple), Amelanchier spp.(serviceberry), and Gleditsia 210 triacanthos (honey locust) were found in sufficient quantities to be included in two-way 211 ANOVA analysis.Results indicated that DBH was significantly affected by stock (p [1,15] =0.020) 212 and species (p [2,15] =0.029).Percent canopy cover was significantly affected by species 213 (p [2,15] <0.0001).Tukey tests indicated that with regard to DBH, Amelanchier spp. vs. A. rubrum 214 (p<0.0001) as well as G. triacanthos vs. Amelanchier (p<0.0001) were significantly different;(p=0.022)also had lower percents live crown.Age, stock, species, pit cut size, wound presence, 251 site type and street traffic accounted for 25% of varience in percent live crown (

120
Penn State Cooperative Extension.The training program covers tree planting, identification, benefits, and maintenance.The fact that TreeVitalize represents both public and private 122 organizations, as well as community groups makes it an ideal urban forestry program for study.

Table 1 -
Table1).Higher mortality was found along commercial corridors (the highest traffic intensity).

Table 4 -
Buckstrup and Bassuk's (2000) recorded in older, B&B, street, yard, Platanus x acerifolia Figure 4 -Regression trees illustrate significant interactions between factors affecting growth 263 measures.They can be read from the top down; longer branches indicate more significant trends.For example, regarding percent live crown, trees under the age of 2.5 years, that are bare root, aThe main objective of this study was to investigate the influence of stock (B&B vs. bare This study cannot provide insight into the comparative rates of growth; however, it does 280 highlight that B&B trees currently have larger DBH measures and fuller crowns on average.Significance of stock on DBH dissipates with age; such that DBH of older trees becomes more 282 correlated with species (Figure4).This supportsBuckstrup and Bassuk's (2000)finding that it encouraged in order to support future research.A follow-up study on the same trees measured for 330 this study would generate valuable information about growth and survival over time.
264 271 root) on survival and growth of urban street trees planted by community groups.Both bare root 272 and B&B displayed very high survival rates, with no significant difference across stock.This 273 substantiates findings by Vanstone and Ronald (1981), Buckstrup and Bassuk (2000), and Anella 274 (2008) within the context of community-based plantings.275With regard to growth measures, B&B trees were found consistently to have higher 276 DBH, percent canopy cover, and percent live crown values.This was expected because B&B 277 trees are regularly planted larger than their bare root counterparts.Because this study was based 278 on previously planted trees, size at which they were planted was not standardized across stocks.279281